AUK Finances and Website Project was: AUK Chairman Statement

[frankly frankie]

What is being produced is a tool to manage the activities that we as an organisation take part in. That could be done as a bestpoke app you down load to your phone, or a spread sheet, or as we have chosen, a website. It is a tool, nothing more nothing less.

For a distributed operation over a wide physical area, a website has been a convenient lingua franca for the last 25 years.

[quixoticgeek]

What is being produced is a tool to manage the activities that we as an organisation take part in. That could be done as a bestpoke app you down load to your phone, or a spread sheet, or as we have chosen, a website. It is a tool, nothing more nothing less.

For a distributed operation over a wide physical area, a website has been a convenient lingua franca for the last 25 years.

Yep. But ultimately it's a tool. Unfortunately a website just looks like a website to most people, so when people see that they can get a pretty looking website from square space for the price of a round of drinks, trying to tell them that our new site is going to cost them the same as a house in many parts of the country is a bloody hard sell.

J

[Jaded]

Surely it is not a website, but a database accessed by a website.

Websites are as universal as you can get in the internet. Apps, spreadsheets, all have at least one more hurdle to leap before being accessible to many. They will never be as accessible to as many as a website.

[bludger]

This subject will scare off a lot of people who don't feel equipped with the knowledge to make comments. But it can't be right as QG says that something which (it seems to me) does broadly the same thing as RiderHQ and Eventbrite has come to the same cost as the price of a house. I expect that a major culprit is the requirement that the system migrate over old records and calculate people's awards e.g. the Brevet ribbons etc, plus the PBP qualification gubbins.

I would put that it may be a more future-proof solution to actually go and offer Cycling UK the opportunity to manage the Audax UK IT stuff, perhaps operating Audax UK as a microsite off the back of cyclinguk.org. We are cycling uk affiliated anyway, especially in the wake of new trends towards long distance/non 'road race' cycling it might well fit in with their 'core business' and be a good fit. Then Audax UK would become less reliant on the (much-appreciated) time and energy of the volunteers - or rather, the volunteers might enjoy being involved with side of things directly concerned with promoting the rides and the disciplien more widely among UK cyclists.

This would obviously be a very big change to the nature of AUK and goes beyond the scope of 'fixing the website' per se. It would probably go to some kind of big vote.

[FifeingEejit]

What is being produced is a tool to manage the activities that we as an organisation take part in. That could be done as a bestpoke app you down load to your phone, or a spread sheet, or as we have chosen, a website. It is a tool, nothing more nothing less.

For a distributed operation over a wide physical area, a website has been a convenient lingua franca for the last 25 years.

Yep. But ultimately it's a tool. Unfortunately a website just looks like a website to most people, so when people see that they can get a pretty looking website from square space for the price of a round of drinks, trying to tell them that our new site is going to cost them the same as a house in many parts of the country is a bloody hard sell.

J

That's something that's far too often forgotten, everything that "IT" does needs to fulfill a business need.
If you don't have a business need, there's no magic wand that will create one. (that's not necessarily a dig at AUK).


AUK has decided that it's business needs include various things on top of the standard ACP representations.
Oddly enough I believe there's people who have disagreed with that since day one.

On an earlier point, no matter what you pick as the technical stack there's always going to be lock in.
and oddly enough there's always going to be people evangelizing using another stack even long after the decision is made.


The stopping and asking for directions seems to be because people have disagreed with the destination since setting off on the journey.

[FifeingEejit]

This subject will scare off a lot of people who don't feel equipped with the knowledge to make comments. But it can't be right as QG says that something which (it seems to me) does broadly the same thing as RiderHQ and Eventbrite has come to the same cost as the price of a house. I expect that a major culprit is the requirement that the system migrate over old records and calculate people's awards e.g. the Brevet ribbons etc, plus the PBP qualification gubbins.

You mean the non-standard bespoke requirements that needs a load of custom work done to provide?
But neither of them have been delivered yet, just apparently the standard website stuff.

We've got this "great" website here, it does all sorts of stuff for the business that you only find in this sector, it's a multi-million pound a year contract.
(For those with interest in the matter it's our PAS)

[hellymedic]

This subject will scare off a lot of people who don't feel equipped with the knowledge to make comments. But it can't be right as QG says that something which (it seems to me) does broadly the same thing as RiderHQ and Eventbrite has come to the same cost as the price of a house. I expect that a major culprit is the requirement that the system migrate over old records and calculate people's awards e.g. the Brevet ribbons etc, plus the PBP qualification gubbins.

I would put that it may be a more future-proof solution to actually go and offer Cycling UK the opportunity to manage the Audax UK IT stuff, perhaps operating Audax UK as a microsite off the back of cyclinguk.org. We are cycling uk affiliated anyway, especially in the wake of new trends towards long distance/non 'road race' cycling it might well fit in with their 'core business' and be a good fit. Then Audax UK would become less reliant on the (much-appreciated) time and energy of the volunteers - or rather, the volunteers might enjoy being involved with side of things directly concerned with promoting the rides and the disciplien more widely among UK cyclists.

This would obviously be a very big change to the nature of AUK and goes beyond the scope of 'fixing the website' per se. It would probably go to some kind of big vote.

I don't think the history of Cycling UK's website development and issues makes it wise for AUK to take any leaves from that book...

[bludger]

This subject will scare off a lot of people who don't feel equipped with the knowledge to make comments. But it can't be right as QG says that something which (it seems to me) does broadly the same thing as RiderHQ and Eventbrite has come to the same cost as the price of a house. I expect that a major culprit is the requirement that the system migrate over old records and calculate people's awards e.g. the Brevet ribbons etc, plus the PBP qualification gubbins.

You mean the non-standard bespoke requirements that needs a load of custom work done to provide?

There's requirements and then there's requirements.  If the 'move' is done to coincide with a new Audax season then it's only really RRTY that gets missed and riders who are keen could just be asked to keep their validation files/documents on hand for the duration of the move, and then to validate their RRTY records once the new system is online.

I am not kidding that Cycling UK haven't made their own errors but the fact is that they have got full-time staff who can give this the attention it deserves. Of course maybe they'd take one look at this whole thing and say 'not with a barge pole old chap'.

[grams]

The custom development so far has been displaying calendar event listings and member results, pulled in from the old database, all of which is done "live".

The problems with saying "just use a clean slate new database" are:
- you still need to write the code to pull and display data from your shiny new database
- the tools to insert and update data (calendar events and results)  into the new database don't yet exist.

I suspect the latter could end up being a monster project if it attempts to recreate what's already there.

These problems (and costs) are all inherent to the "rewrite everything from scratch" approach, however you dice it. The solutions around are either:
- Find a way to live with the old system.
- Build something much much simpler that doesn't attempt to recreate what exists on AUKweb.

[bludger]

Partly why I thought of cycling UK is that they do have functioning ride calendars and so on ( see https://www.cyclinguk.org/events-listing ), and so I presume it is fairly simple to create/manage events. My thinking is that a Cycling UK microsite could use the same behind the scenes gubbins but render the output on an 'audax uk' microsite so people just looking for audaxes,retaining the audax UK brand but off the back of cycling UK machinery.

[citoyen]

Partly why I thought of cycling UK is that they do have functioning ride calendars and so on ( see https://www.cyclinguk.org/events-listing ), and so I presume it is fairly simple to create/manage events.

Looking at that link, what Cycling UK offers is basic listings, which you can do easily with a flat database and simple php-driven web page. What AUK does is event and member management which presumably requires a vastly more complex back end. The two are not remotely comparable except on a very superficial level.

[FifeingEejit]

These problems (and costs) are all inherent to the "rewrite everything from scratch" approach, however you dice it. The solutions around are either:
- Find a way to live with the old system.
- Build something much much simpler that doesn't attempt to recreate what exists on AUKweb.

The first there is disagreement over but the "owner" of the existing code seems to be against, which gives extra weight to not doing that.
The later, what would be viable to remove from the AukWeb set? And without replacement?

Stopping doing event management and membership management aren't exactly feasible, so some form of replacement is still needed
Which leaves finish lists and the functions that hang off them.  But finish lists of some form are needed and the functions that hang off of them are surely fairly trivial queries but are also the most visible part of the system to many members after the event management even though they're really just added value.

[bludger]

Partly why I thought of cycling UK is that they do have functioning ride calendars and so on ( see https://www.cyclinguk.org/events-listing ), and so I presume it is fairly simple to create/manage events.

Looking at that link, what Cycling UK offers is basic listings, which you can do easily with a flat database and simple php-driven web page. What AUK does is event and member management which presumably requires a vastly more complex back end. The two are not remotely comparable except on a very superficial level.

The Cycling UK website does do member management. You log in using your Cycling UK account.

I expect the cyclingUK Drupal website could provide event functionality on the basis of this article https://www.adaptive.co.uk/blog/drupal-8-event-registration-platform-developed-independent-age I think the only reason cyclinguk's website doesn't do more is that their local groups like to retain direct control of their own comms, see e.g. https://www.centrallondonctc.org.uk/

CyclingUK also have their 'holidays' website which is worth looking at from an event management front see e.g https://www.cyclingholidays.org/european-cycling-holidays/details.php?id=466

[Martin]

I'll ask again, after AUK and its volunteers have done all they can to migrate over to the new site, how much will remain on the old site? and more importantly how long will the old site last

asking again

[jiberjaber]

I know this may sound odd, but why not ask these questions over on the actual AUK forum itself, asking here is all well and good, but I suspect the people with the answers might only answer it under their own forum not here on a '3rd party' one...

Seems there's a lot of ideas etc here and some valid points which may need explanation - but if it's not going to be entertained because of the location it's being posted on, then it's all a bit of an academic exercise.

[frankly frankie]

The current thinking seems to be of "get Phase 2 done, then take stock" - but that doesn't make a whole lot of sense to me, because "Phase 2 done" is not an acceptable end point to anybody.  From there you either have to press on, or go back, and it seems pretty clear that going back is not an option that anyone with clout in this matter is prepared to contemplate.  I estimate that pressing on would certainly take another 2 years to complete though.  And probably some frills and niceties in the current AUK operation will have to be ditched along the way, and TBH they will be no great loss.

I expect that a major culprit is the requirement that the system migrate over old records and calculate people's awards e.g. the Brevet ribbons etc, plus the PBP qualification gubbins.

It's true that there is a bit of a hang-up over the prospect (in phase 3 presumably) of maintaining compatibility with the legacy results data that has accrued over the last 20 years - I keep telling them to just ignore it, just concentrate on day-to-day running without any eye to the old data.  As far as I'm concerned, everything from 1999 to 'last season' can just be zipped up into a file of about 12Mb and dumped on the cloud somewhere, job done.  Some future data archeologist will find a way to display it nicely.  And everything prior to 1999 we have as scanned images of paper results sheets, going all the way back to 1976.  These are not currently accessible online but it would be nice if they were.

[bludger]

I am very romantic about the idea of 'halls of fame', data of achievements that go back for years, it is one of the USPs of audax versus 'just going for a ride' but we have to be ruthless here, this is huge sums of cash on the line.

[Jaded]

I know this may sound odd, but why not ask these questions over on the actual AUK forum itself, asking here is all well and good, but I suspect the people with the answers might only answer it under their own forum not here on a '3rd party' one...

Seems there's a lot of ideas etc here and some valid points which may need explanation - but if it's not going to be entertained because of the location it's being posted on, then it's all a bit of an academic exercise.

That approach just fuels the conspiracy fire 

[S2L]

  As far as I'm concerned, everything from 1999 to 'last season' can just be zipped up into a file of about 12Mb and dumped on the cloud somewhere, job done.  Some future data archeologist will find a way to display it nicely.  And everything prior to 1999 we have as scanned images of paper results sheets, going all the way back to 1976.  These are not currently accessible online but it would be nice if they were.

It would probably be cheaper to have those results chiselled on a granite monument to display in the centre of Meriden, just  by the cyclist memorial, than paying for this bespoke website.

[S2L]

I know this may sound odd, but why not ask these questions over on the actual AUK forum itself, asking here is all well and good, but I suspect the people with the answers might only answer it under their own forum not here on a '3rd party' one...

It seems to me questions are not answered, neither here, nor there

[rob]

I am very romantic about the idea of 'halls of fame', data of achievements that go back for years, it is one of the USPs of audax versus 'just going for a ride' but we have to be ruthless here, this is huge sums of cash on the line.

Having been a member for 25+ years I still like this.

The Ultra Randonneur list is up to date but the PBP list isn't.

[Kim]

It's true that there is a bit of a hang-up over the prospect (in phase 3 presumably) of maintaining compatibility with the legacy results data that has accrued over the last 20 years - I keep telling them to just ignore it, just concentrate on day-to-day running without any eye to the old data.  As far as I'm concerned, everything from 1999 to 'last season' can just be zipped up into a file of about 12Mb and dumped on the cloud somewhere, job done.  Some future data archeologist will find a way to display it nicely.  And everything prior to 1999 we have as scanned images of paper results sheets, going all the way back to 1976.  These are not currently accessible online but it would be nice if they were.

Indeed.  Importing historical results data into the new system seems eminently suited to a volunteers-in-their-spare-time project.

[FifeingEejit]

It's true that there is a bit of a hang-up over the prospect (in phase 3 presumably) of maintaining compatibility with the legacy results data that has accrued over the last 20 years - I keep telling them to just ignore it, just concentrate on day-to-day running without any eye to the old data.  As far as I'm concerned, everything from 1999 to 'last season' can just be zipped up into a file of about 12Mb and dumped on the cloud somewhere, job done.  Some future data archeologist will find a way to display it nicely.  And everything prior to 1999 we have as scanned images of paper results sheets, going all the way back to 1976.  These are not currently accessible online but it would be nice if they were.

Indeed.  Importing historical results data into the new system seems eminently suited to a volunteers-in-their-spare-time project.

Aye, depending on how different the data is it will either be an easy task to migrate with ETL processes or a painstaking manual nightmare.

Data migration for the system I mentioned earlier was rekeying patients currently in hospital and the portal system reading both for a couple of years.
The cost of doing anything else was astronomical.

Sent from my BKL-L09 using Tapatalk

[Greenbank]

I expect that a major culprit is the requirement that the system migrate over old records and calculate people's awards e.g. the Brevet ribbons etc, plus the PBP qualification gubbins.

It's true that there is a bit of a hang-up over the prospect (in phase 3 presumably) of maintaining compatibility with the legacy results data that has accrued over the last 20 years - I keep telling them to just ignore it, just concentrate on day-to-day running without any eye to the old data.  As far as I'm concerned, everything from 1999 to 'last season' can just be zipped up into a file of about 12Mb and dumped on the cloud somewhere, job done.  Some future data archeologist will find a way to display it nicely.  And everything prior to 1999 we have as scanned images of paper results sheets, going all the way back to 1976.  These are not currently accessible online but it would be nice if they were.

I can't see old data being as huge a problem as that.

If the new system is being designed to handle and display results for rides from its inception forward then migrating the old results into the new format shouldn't be too hard at all. The only reason it could be difficult is if the new format is unnecessarily strict or byzantine. And that only happens if the system is designed to satisfy just the current phase(s) rather than considering the possible future phases. (Can you guess what I think has been going on...)

An awful lot of the awards simply rely on knowing that rider X did a Y00km BR/BRM ride on date Z. Even the more complicated things like RTTY calculations (where a ride might be counted in the next month, or Jan/Feb shenanigans of however many years ago) can be all but automated.

Or a suitable abstraction layer means you don't have to care whether the ride data is coming from the new latest-and-greatest results tables, or from a simple table of legacy ride results.

The majority of awards are vaguely formulaic and could be treated as data themselves and handled by a single algorithm, but quite a few of the awards will require bespoke code to implement and this is true whether you have the legacy data imported somehow or not.

[FifeingEejit]

I expect that a major culprit is the requirement that the system migrate over old records and calculate people's awards e.g. the Brevet ribbons etc, plus the PBP qualification gubbins.

It's true that there is a bit of a hang-up over the prospect (in phase 3 presumably) of maintaining compatibility with the legacy results data that has accrued over the last 20 years - I keep telling them to just ignore it, just concentrate on day-to-day running without any eye to the old data.  As far as I'm concerned, everything from 1999 to 'last season' can just be zipped up into a file of about 12Mb and dumped on the cloud somewhere, job done.  Some future data archeologist will find a way to display it nicely.  And everything prior to 1999 we have as scanned images of paper results sheets, going all the way back to 1976.  These are not currently accessible online but it would be nice if they were.

I can't see old data being as huge a problem as that.

If the new system is being designed to handle and display results for rides from its inception forward then migrating the old results into the new format shouldn't be too hard at all. The only reason it could be difficult is if the new format is unnecessarily strict or byzantine. And that only happens if the system is designed to satisfy just the current phase(s) rather than considering the possible future phases. (Can you guess what I think has been going on...)

An awful lot of the awards simply rely on knowing that rider X did a Y00km BR/BRM ride on date Z. Even the more complicated things like RTTY calculations (where a ride might be counted in the next month, or Jan/Feb shenanigans of however many years ago) can be all but automated.

The majority of awards are vaguely formulaic and could be treated as data themselves and handled by a single algorithm, but quite a few of the awards will require bespoke code to implement and this is true whether you have the legacy data imported somehow or not.

I was working on a way to figure out what's what for formulaic processing for screen scraped data, I found I had to manually turn the points Into distance for an event, not difficult... Until you need to add on ECEs, you can't rely on the E in the homologation number.

From that and previous comments about event types I suspect the legacy data being wooly about what's what is more likely to be problem if the new data structure is more explicit on combining events with their extension.

I got bored after than and went out for a bike ride and never went back to it so I'm still using my spread sheet.

Sent from my BKL-L09 using Tapatalk